Guide

The ins and outs of Rocket, in detail.

Deploying

This section describes deploying Rocket applications to production. It provides a general overview as well as concrete, reusable examples for common deployment scenarios including self-managed deployments, containerization, and fully-managed deployments.

Rocket does not endorse or prefer any particular tools or services.

Rocket does not endorse or prefer any specific tools or services mentioned in this guide. They are mentioned in exposition only. Rocket is agnostic to its deployment environment.

Overview

For any deployment, it's important to keep in mind:

  1. Configuration

    Minimally, Rocket will need to be configured to listen on the correct port and address, typically port 80 or 8080 and address 0.0.0.0. Your deployment environment may have different requirements. Recall that by default, you can set the address and port via the environment variables ROCKET_ADDRESS and ROCKET_PORT as well as through many other means.

  2. Asset Bundling

    If your application serves assets or leverages templates, you may need to bundle those assets with your application binary. For example, if you serve static assets from the ./static directory and enable templates, you'll need to ensure that those directories are present and in the current working directory that your application binary starts in.

    1
2
3
4
5
6
7
8
9
    use rocket::fs::FileServer;
use rocket_dyn_templates::Template;

#[launch]
fn rocket() -> _ {
    rocket::build()
        .mount("/", FileServer::from("./static"))
        .attach(Template::fairing())
}

    For the application above, assuming the template_dir configuration parameter hasn't been changed, you'll need to ensure that the static and templates directories are placed in the current working directory that the application will start in. Otherwise, Rocket will refuse to launch.

  3. Load Balancing

    Rocket does not yet have robust support for DDoS mitigation, so a production deployment will require placing Rocket behind a load balancer or reverse proxy that does. If you are deploying your Rocket application to managed environments such as Kubernetes, Heroku, or Google Cloud Run, this will be handled for you automatically. However, if you're deploying to a self-managed environment such as a VPS, we recommend placing Rocket behind a mature reverse proxy such as HAProxy or NGINX.

  4. Service Management

    As your application matures, you'll need to deploy updated versions to your production environment, stopping the existing application and starting the new one in its place. In a managed environment, you can likely rely on the environment to provide these mechanisms. In a self-managed environment, using a service manager like systemd, in addition to a reverse proxy, is recommended.

    In either case, it's important to know that once a Rocket application has started, it will run until graceful shutdown is initiated. Your application should leverage Rocket's graceful shutdown mechanisms such as the Shutdown future and shutdown fairings to clean-up resources before terminating. You should also ensure that the graceful shutdown configuration is aligned with your environment. For example, Kubernetes issues a SIGTERM signal to initiate termination which Rocket listens for by default, but other environments may send other signals which you might need to enable as triggers.

The following section addresses these concerns and more for common deployment scenarios.

Common Scenarios

Self-Managed

In a self-managed environment, you are typically responsible for all facets of deployment and service management. In exchange, a self-managed environment typically incurs the lowest financial cost.

You must decide whether you manage your Rocket application directly, by installing, configuring, and running a service manager, load balancer, and Rocket application, or indirectly by installing and configuring an application management service like kubernetes, k3s, or dokku. Because indirect self-management typically revolves around containerization, covered in the next section, we focus on direct self-management here.

Our recommendation for a direct self-managed deployment is to:

Containerization

In a containerization environment, you are responsible for writing a Dockerfile or Containerfile which you provide to an application platform. The platform may be self-managed, as with k3s or dokku, or fully-managed, as with Google Cloud Run or Heroku.

Below you'll find an example of a Dockerfile that:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
FROM docker.io/rust:1-slim-bookworm AS build

## cargo package name: customize here or provide via --build-arg
ARG pkg=rocket-app

WORKDIR /build

COPY . .

RUN --mount=type=cache,target=/build/target \
    --mount=type=cache,target=/usr/local/cargo/registry \
    --mount=type=cache,target=/usr/local/cargo/git \
    set -eux; \
    cargo build --release; \
    objcopy --compress-debug-sections target/release/$pkg ./main

################################################################################

FROM docker.io/debian:bookworm-slim

WORKDIR /app

## copy the main binary
COPY --from=build /build/main ./

## copy runtime assets which may or may not exist
COPY --from=build /build/Rocket.tom[l] ./static
COPY --from=build /build/stati[c] ./static
COPY --from=build /build/template[s] ./templates

## ensure the container listens globally on port 8080
ENV ROCKET_ADDRESS=0.0.0.0
ENV ROCKET_PORT=8080

CMD ./main

You will need to modify the pkg ARG or provide it via the command-line:

1
docker build --build-arg pkg=cargo_package_name -t app  .

You may also need to make the following changes:

Finally, we recommend the following .dockerignore file to avoid copying unnecessary artifacts:

1
2
3
4
target
.cargo
**/*.sh
**/*.tar.gz

Fully-Managed

In a fully-managed environment, you provide a service with your source code and instructions on how to build and run your application. The Dockerfile in the containerization section, coupled with a configuration file that instructs the service to build it, may be one such example.

Because the specifics on deploying to a fully-managed environment depend on the environment, we provide only the following general guidelines:

◄ Configuration
Pastebin Tutorial ►